arXiv subjects

Samir M. Perlaza

Publications and source records attributed to Samir M. Perlaza.

At least 37 records · Page 2Linked to original sources

arXiv2023

Generalization Analysis of Machine Learning Algorithms via the Worst-Case Data-Generating Probability Measure

In this paper, the worst-case probability measure over the data is introduced as a tool for characterizing the generalization capabilities of machine learning algorithms. More specifically, the worst-case probability measure is a Gibbs probability measure and the unique solution to the maximization of the expected loss under a relative entropy constraint with respect to a reference probability measure. Fundamental generalization metrics, such as the sensitivity of the expected loss, the sensitivity of the empirical risk, and the generalization gap are shown to have closed-form expressions involving the worst-case data-generating probability measure. Existing results for the Gibbs algorithm, such as characterizing the generalization gap as a sum of mutual information and lautum information, up to a constant factor, are recovered. A novel parallel is established between the worst-case data-generating probability measure and the Gibbs algorithm. Specifically, the Gibbs probability measure is identified as a fundamental commonality of the model space and the data space for machine learning algorithms.

cs.LG↗

arXiv2023

On the Validation of Gibbs Algorithms: Training Datasets, Test Datasets and their Aggregation

The dependence on training data of the Gibbs algorithm (GA) is analytically characterized. By adopting the expected empirical risk as the performance metric, the sensitivity of the GA is obtained in closed form. In this case, sensitivity is the performance difference with respect to an arbitrary alternative algorithm. This description enables the development of explicit expressions involving the training errors and test errors of GAs trained with different datasets. Using these tools, dataset aggregation is studied and different figures of merit to evaluate the generalization capabilities of GAs are introduced. For particular sizes of such datasets and parameters of the GAs, a connection between Jeffrey's divergence, training and test errors is established.

cs.LG↗

arXiv2023

Analysis of the Relative Entropy Asymmetry in the Regularization of Empirical Risk Minimization

The effect of the relative entropy asymmetry is analyzed in the empirical risk minimization with relative entropy regularization (ERM-RER) problem. A novel regularization is introduced, coined Type-II regularization, that allows for solutions to the ERM-RER problem with a support that extends outside the support of the reference measure. The solution to the new ERM-RER Type-II problem is analytically characterized in terms of the Radon-Nikodym derivative of the reference measure with respect to the solution. The analysis of the solution unveils the following properties of relative entropy when it acts as a regularizer in the ERM-RER problem: i) relative entropy forces the support of the Type-II solution to collapse into the support of the reference measure, which introduces a strong inductive bias that dominates the evidence provided by the training data; ii) Type-II regularization is equivalent to classical relative entropy regularization with an appropriate transformation of the empirical risk function. Closed-form expressions of the expected empirical risk as a function of the regularization parameters are provided.

cs.IT↗

arXiv2023

Information-Energy Regions in the Finite Block-Length Regime with Finite Channel Inputs

This paper characterizes the trade-offs between information and energy transmission over an additive white Gaussian noise channel in the finite block-length regime with finite sets of channel input symbols. These trade-offs are characterized using impossibility and achievability bounds on the information transmission rate, energy transmission rate, decoding error probability (DEP) and energy outage probability (EOP) for a finite block-length code. Given a set of channel input symbols, the impossibility results identify the tuples of information rate, energy rate, DEP and EOP that cannot be achieved by any code using the given set of channel inputs. A novel method for constructing a family of codes that satisfy a target information rate, energy rate, DEP and EOP is also proposed. The achievability bounds identify the set of tuples of information rate, energy rate, DEP and EOP that can be simultaneously achieved by the constructed family of codes. The proposed construction matches the impossibility bounds for the information rate, energy rate, and the EOP. However, for a given information rate, energy rate and EOP, the achieved DEP is higher than the impossibility bound due to the choice of the decoding sets made during the code construction.

cs.IT↗

arXiv2023

$2 \times 2$ Zero-Sum Games with Commitments and Noisy Observations

In this paper, $2\times2$ zero-sum games are studied under the following assumptions: $(1)$ One of the players (the leader) commits to choose its actions by sampling a given probability measure (strategy); $(2)$ The leader announces its action, which is observed by its opponent (the follower) through a binary channel; and $(3)$ the follower chooses its strategy based on the knowledge of the leader's strategy and the noisy observation of the leader's action. Under these conditions, the equilibrium is shown to always exist. Interestingly, even subject to noise, observing the actions of the leader is shown to be either beneficial or immaterial for the follower. More specifically, the payoff at the equilibrium of this game is upper bounded by the payoff at the Stackelberg equilibrium (SE) in pure strategies; and lower bounded by the payoff at the Nash equilibrium, which is equivalent to the SE in mixed strategies.Finally, necessary and sufficient conditions for observing the payoff at equilibrium to be equal to its lower bound are presented. Sufficient conditions for the payoff at equilibrium to be equal to its upper bound are also presented.

cs.GT↗

arXiv2022

Empirical Risk Minimization with Relative Entropy Regularization: Optimality and Sensitivity Analysis

The optimality and sensitivity of the empirical risk minimization problem with relative entropy regularization (ERM-RER) are investigated for the case in which the reference is a sigma-finite measure instead of a probability measure. This generalization allows for a larger degree of flexibility in the incorporation of prior knowledge over the set of models. In this setting, the interplay of the regularization parameter, the reference measure, the risk function, and the empirical risk induced by the solution of the ERM-RER problem is characterized. This characterization yields necessary and sufficient conditions for the existence of a regularization parameter that achieves an arbitrarily small empirical risk with arbitrarily high probability. The sensitivity of the expected empirical risk to deviations from the solution of the ERM-RER problem is studied. The sensitivity is then used to provide upper and lower bounds on the expected empirical risk. Moreover, it is shown that the expectation of the sensitivity is upper bounded, up to a constant factor, by the square root of the lautum information between the models and the datasets.

cs.LG↗

arXiv2022

An information theoretic vulnerability metric for data integrity attacks on smart grids

A novel metric that describes the vulnerability of the measurements in power systems to data integrity attacks is proposed. The new metric, coined vulnerability index (VuIx), leverages information theoretic measures to assess the attack effect on the fundamental limits of the disruption and detection tradeoff. The result of computing the VuIx of the measurements in the system yields an ordering of their vulnerability based on the level of exposure to data integrity attacks. This new framework is used to assess the measurement vulnerability of IEEE 9-bus and 30-bus test systems and it is observed that power injection measurements are overwhelmingly more vulnerable to data integrity attacks than power flow measurements. A detailed numerical evaluation of the VuIx values for IEEE test systems is provided.

eess.SY↗

arXiv2022

Stealth Data Injection Attacks with Sparsity Constraints

Sparse stealth attack constructions that minimize the mutual information between the state variables and the observations are proposed. The attack construction is formulated as the design of a multivariate Gaussian distribution that aims to minimize the mutual information while limiting the Kullback-Leibler divergence between the distribution of the observations under attack and the distribution of the observations without attack. The sparsity constraint is incorporated as a support constraint of the attack distribution. Two heuristic greedy algorithms for the attack construction are proposed. The first algorithm assumes that the attack vector consists of independent entries, and therefore, requires no communication between different attacked locations. The second algorithm considers correlation between the attack vector entries which results in better attack performance at the expense of coordination between different locations. We numerically evaluate the performance of the proposed attack constructions on IEEE test systems and show that it is feasible to construct stealth attacks that generate significant disruption with a low number of compromised sensors.

eess.SY↗

arXiv2022

Information Theoretic Data Injection Attacks with Sparsity Constraints

Information theoretic sparse attacks that minimize simultaneously the information obtained by the operator and the probability of detection are studied in a Bayesian state estimation setting. The attack construction is formulated as an optimization problem that aims to minimize the mutual information between the state variables and the observations while guaranteeing the stealth of the attack. Stealth is described in terms of the Kullback-Leibler (KL) divergence between the distributions of the observations under attack and without attack. To overcome the difficulty posed by the combinatorial nature of a sparse attack construction, the attack case in which only one sensor is compromised is analytically solved first. The insight generated in this case is then used to propose a greedy algorithm that constructs random sparse attacks. The performance of the proposed attack is evaluated in the IEEE 30 Bus Test Case.

eess.SY↗

arXiv2022

Power Injection Measurements are more Vulnerable to Data Integrity Attacks than Power Flow Measurements

A novel metric that describes the vulnerability of the measurements in power system to data integrity attacks is proposed. The new metric, coined vulnerability index (VuIx), leverages information theoretic measures to assess the attack effect on the fundamental limits of the disruption and detection tradeoff. The result of computing the VuIx of the measurements in the system yields an ordering of the measurements vulnerability based on the level of exposure to data integrity attacks. This new framework is used to assess the measurements vulnerability of IEEE test systems and it is observed that power injection measurements are overwhelmingly more vulnerable to data integrity attacks than power flow measurements. A detailed numerical evaluation of the VuIx values for IEEE test systems is provided.

eess.SY↗

arXiv2022

FEAT: Fair Coordinated Iterative Water-Filling Algorithm

In this paper, we consider a perfect coordinated water-filling game, where each user transmits solely on a given carrier. The main goal of the proposed algorithm (which we call FEAT) is to get close to the optimal, while keeping a decent level of fairness. The key idea within FEAT is to minimize the ratio between the best and the worst utilities of the users. This is done by ensuring that, at each iteration (channel assignment), a user is satisfied with this assignment as long as he does not loose much more than other users in the system. It has been shown that FEAT outperforms most related algorithms in many aspects, especially in interference-limited systems. Indeed, with FEAT we can ensure a near-optimal, fair and energy efficient solution with low computational complexity. In terms of robustness, it turns out that the balance between being nearly globally optimal and good from individual point of view seems hard to sustain with a significant number of users. Also notice that, in this regard, global optimality gets less affected than the individual one, which offers hope that such an accurate water-filling algorithm can be designed around competition in interference-limited systems.

cs.IT↗

arXiv2022

Achievable Information-Energy Region in the Finite Block-Length Regime with Finite Constellations

This paper characterizes an achievable information-energy region of simultaneous information and energy transmission over an additive white Gaussian noise channel. This analysis is performed in the finite block-length regime with finite constellations. More specifically, a method for constructing a family of codes is proposed and the set of achievable tuples of information rate, energy rate, decoding error probability (DEP) and energy outage probability (EOP) is characterized. Using existing converse results, it is shown that the construction is information rate, energy rate, and EOP optimal. The achieved DEP is, however, sub-optimal.

cs.IT↗

arXiv2021

Simultaneous Information and Energy Transmission with Finite Constellations

In this paper, the fundamental limits on the rates at which information and energy can be simultaneously transmitted over an additive white Gaussian noise channel are studied under the following assumptions: $(a)$ the channel is memoryless; $(b)$ the number of channel input symbols (constellation size) and block length are finite; and $(c)$ the decoding error probability (DEP) and the energy outage probability (EOP) are bounded away from zero. In particular, it is shown that the limits on the maximum information and energy transmission rates; and the minimum DEP and EOP, are essentially set by the type induced by the code used to perform the transmission. That is, the empirical frequency with which each channel input symbol appears in the codewords. Using this observation, guidelines for optimal constellation design for simultaneous energy and information transmission are presented.

cs.IT↗

arXiv2021

Data-Injection Attacks

In this chapter we review some of the basic attack constructions that exploit a stochastic description of the state variables. We pose the state estimation problem in a Bayesian setting and cast the bad data detection procedure as a Bayesian hypothesis testing problem. This revised detection framework provides the benchmark for the attack detection problem that limits the achievable attack disruption. Indeed, the trade-off between the impact of the attack, in terms of disruption to the state estimator, and the probability of attack detection is analytically characterized within this Bayesian attack setting. We then generalize the attack construction by considering information-theoretic measures that place fundamental limits to a broad class of detection, estimation, and learning techniques. Because the attack constructions proposed in this chapter rely on the attacker having access to the statistical structure of the random process describing the state variables, we conclude by studying the impact of imperfect statistics on the attack performance. Specifically, we study the attack performance as a function of the size of the training data set that is available to the attacker to estimate the second-order statistics of the state variables.

eess.SY↗

arXiv2020

Stealth Attacks on the Smart Grid

Random attacks that jointly minimize the amount of information acquired by the operator about the state of the grid and the probability of attack detection are presented. The attacks minimize the information acquired by the operator by minimizing the mutual information between the observations and the state variables describing the grid. Simultaneously, the attacker aims to minimize the probability of attack detection by minimizing the Kullback-Leibler (KL) divergence between the distribution when the attack is present and the distribution under normal operation. The resulting cost function is the weighted sum of the mutual information and the KL divergence mentioned above. The tradeoff between the probability of attack detection and the reduction of mutual information is governed by the weighting parameter on the KL divergence term in the cost function. The probability of attack detection is evaluated as a function of the weighting parameter. A sufficient condition on the weighting parameter is given for achieving an arbitrarily small probability of attack detection. The attack performance is numerically assessed on the IEEE 30-Bus and 118-Bus test systems.

cs.IT↗

arXiv2018

Approximate Nash Region of the Gaussian Interference Channel with Noisy Output Feedback

In this paper, an achievable $η$-Nash equilibrium ($η$-NE) region for the two-user Gaussian interference channel with noisy channel-output feedback is presented for all $η\geqslant 1$. This result is obtained in the scenario in which each transmitter-receiver pair chooses its own transmit-receive configuration in order to maximize its own individual information transmission rate. At an $η$-NE, any unilateral deviation by either of the pairs does not increase the corresponding individual rate by more than $η$ bits per channel use.

cs.IT↗

arXiv2018

Approximate Capacity Region of the Two-User Gaussian Interference Channel with Noisy Channel-Output Feedback

In this paper, the capacity region of the linear deterministic interference channel with noisy channel-output feedback (LD-IC-NF) is fully characterized. The proof of achievability is based on random coding arguments and rate splitting; block-Markov superposition coding; and backward decoding. The proof of converse reuses some of the existing outer bounds and includes new ones obtained using genie-aided models. Following the insight gained from the analysis of the LD-IC-NF, an achievability region and a converse region for the two-user Gaussian interference channel with noisy channel-output feedback (GIC-NF) are presented. Finally, the achievability region and the converse region are proven to approximate the capacity region of the G-IC-NF to within 4.4 bits.

cs.IT↗

arXiv2017

Decentralized Simultaneous Information and Energy Transmission in K-User Multiple Access Channels

In this paper, the fundamental limits of decentralized simultaneous information and energy transmission in the $K$-user Gaussian multiple access channel (G-MAC), with an arbitrary $K \geqslant 2$ and one non-colocated energy harvester (EH), are fully characterized. The objective of the transmitters is twofold. First, they aim to reliably communicate their message indices to the receiver; and second, to harvest energy at the EH at a rate not less than a minimum rate requirement $b$. The information rates $R_1,\dots,R_K$, in bits per channel use, are measured at the receiver and the energy rate $B$ is measured at an EH. Stability is considered in the sense of an $η$-Nash equilibrium ($η$-NE), with $η> 0$. The main result is a full characterization of the $η$-NE information-energy region, i.e., the set of information-energy rate tuples $(R_1,\dots,R_K,B)$ that are achievable and stable in the G-MAC when: $(a)$ all the transmitters autonomously and independently tune their own transmit configurations seeking to maximize their own information transmission rates $R_1,\dots, R_K$; and $(b)$ all the transmitters jointly guarantee an energy transmission rate $B$ at the EH, such that $B \geqslant b$. Therefore, any rate tuple outside the $η$-NE region is not stable as there always exists at least one transmitter able to increase by at least $η$ bits per channel use its own information transmission rate by updating its own transmit configuration.

cs.IT↗